- Email: [email protected]
XAFS studies on battery materials: Data analysis supported by a chemometric approach
Radiation Physics and Chemistry xxx (xxxx) xxx–xxx
Contents lists available at ScienceDirect
Radiation Physics and Chemistry journal homepage: www.elsevier.com/locate/radphyschem
XAFS studies on battery materials: Data analysis supported by a chemometric approach Marco Giorgettia,∗, Angelo Mullaliua, Paolo Contib a b
Department of Industrial Chemistry “Toso Montanari”, University of Bologna, Viale Risorgimento 4, 40136, Bologna, Italy School of Science, Chemistry Division, University of Camerino, 62032, Camerino, Italy
A R T I C LE I N FO
A B S T R A C T
Keywords: XAFS EXAFS MCR-ALS Li-ion battery Operando
The large amount of data collected during a typical operando X-ray absorption fine structure (XAFS) experiment and the interest to thoroughly investigate and comprehend a redox process occurring in a Li-ion battery need an advanced data analysis approach. In this context, multivariate curve resolution (MCR) has gained attention as chemometric technique that allows the identification of the number of the species and their existence range without any a priori information on the studied system. The call for an advanced data treatment is answered by a joint XAFS-MCR approach, able to unveil the local structural dynamics of the selected species, providing a quick and efficient data interpretation.
1. Introduction The development of reliable, low-cost, sustainable storage of electrical energy generated by sources other than fossil fuels is a global priority. Rechargeable batteries, which are based on high rate intercalation reaction of lithium ions into nano- and micro-structured porous materials, offer efficient electrochemical energy storage (Goodenough and Kim, 2010). The structural and electronic evolution of such hosts is of fundamental importance to design advanced tailored materials, making X-ray absorption fine structure (XAFS) the technique of choice for retrieving this information (Giorgetti, 2013). During a typical operando XAFS experiment, a large amount of data can be acquired, calling for a suitable strategy to data treatment: to this purpose, a joint XAFS-chemometric approach on large datasets allows an efficient and in-depth investigation of the (de)lithiation process. By recording the electrochemical response relatively slowly compared to the spectral data collection, it is possible to build an experimental data matrix (XS,W) where each row is a XAFS spectrum in a very narrow potential range. Multivariate curve resolution refined by alternating least squares (MCR-ALS) provides an additive bilinear model of pure contributions without any preexisting model or a priori information on the system, decomposing the experimental matrix in a product of two matrices to which chemical meaning can be attributed (De Juan et al., 2014; Conti et al., 2010), as it follows (Eqn. (1)):
X s,w = Cs,f Af,w + Es,w ∗
(1)
where CS,F is the column matrix of the concentration profile, AF,W is the row matrix of the X-ray absorption near edge structure (XANES) spectra of pure species and ES,W is the matrix of the residue. A tentative initial value, as close as possible to the final solution, of one of the matrices must be provided for initializing the ALS refinement. A decomposition of the experimental matrix similar to that in Eqn. (1) can be done by Principal component analysis (PCA), however, the two computed matrices do not have physico-chemical meaning, being the main result of PCA the rank of the matrix XS,W that, in the present example, would correspond to the number of pure species. Evolving factor analysis (EFA) (Maeder and Zuberbuehler, 1986) and fixed size window EFA (FSWEFA) (Keller and Massart, 1991) methods (Vandeginste et al., 1998) are preferred in place of PCA because they are able of guessing both the number of pure species and a first estimate of their concentration profiles that are useful to initialize the refinement (Vandeginste et al., 1998; Marassi et al., 2002). Moreover, to obtain a unique and meaningful solution, constraints such as the closure condition and non-negativity (of concentration and/or of pure response) must be applied. A logical tree diagram (Scheme 1) is reported to visualize the steps involved in the MCR-ALS analysis. MCR-ALS technique has been applied in several fields, for instance in chromatography (de Juan and Tauler, 2007), and on in situ XAFS data on catalysis (Cassinelli et al., 2013; Hong et al., 2015). The first example of a joint XAFS-MCR approach dates back to 2010 (Conti et al., 2010), when Conti et al. studied Cu-doped vanadium pentoxide (Cu0.1V2O5) for its attractive properties as insertion host for Li-ions
Corresponding author. E-mail address: [email protected] (M. Giorgetti).
https://doi.org/10.1016/j.radphyschem.2019.04.002 Received 21 September 2018; Received in revised form 14 November 2018; Accepted 1 April 2019 Available online 04 April 2019 0969-806X/ © 2019 Elsevier Ltd. All rights reserved.
Please cite this article as: Marco Giorgetti, Angelo Mullaliu and Paolo Conti, Radiation Physics and Chemistry, https://doi.org/10.1016/j.radphyschem.2019.04.002
Radiation Physics and Chemistry xxx (xxxx) xxx–xxx
M. Giorgetti, et al.
2. Experimental X-ray absorption spectroscopy experiments for FeHCC and CuHCF cathodes were performed at the XAFS beamline of Elettra Sincrotrone in Basovizza (Italy). (Aquilanti et al., 2017). Data collection was carried out during the first charge/discharge cycle in the 1.8 < E < 4.0 V and 2.0 < E < 4.3 V vs. Li+/Li, respectively. The data were recorded in transmission mode either at Fe, Co and/or Cu K-edges. The energies were defined by assigning the first inflection point of the spectra of the metallic foil of Fe, Co and/or Cu to 7112, 7709 and 8979 eV, respectively. An internal reference of Fe, Co and/or Cu foil was used for energy calibration in each scan. Experiments for V2O5 ARG-like were performed at the V K-edge at the National Synchrotron Light Source (NSLS) at Brookhaven National Laboratory using the beamline X11 A (Conti et al., 2010). Data collection was carried out in transmission mode while discharging from open voltage circuit (OCP) value (3.3 V) to 1.5 V vs. Li+/Li. The energy was defined by assigning the first inflection point of the spectra of a vanadium foil. The extended X-ray absorption fine structure (EXAFS) analysis was performed by using the GNXAS package (Filipponi et al., 1995; Filipponi and Di Cicco, 1995) that takes into account the multiple scattering (MS) theory. The theoretical signal is calculated ab initio and contains the relevant two-body γ(2), three-body γ(3), and four-body γ(4) MS terms, (Giorgetti et al., 1997). A typical structure of hexacyanoferrates (Giorgetti and Berrettoni, 2008) was used as structural model for the FeHCC and CuHCF MS signals calculation. The MCR-ALS analysis was performed by taking into account the XANES portion of the XAFS spectrum exclusively. The Unscrambler (Camo Software) and homemade routines based on Parvus (Forina et al., 2008) was used in data treatment.
Scheme 1. MCR-ALS analysis diagram.
batteries. Although the compound was known since years among the scientific community, the actual reaction mechanism occurring during charge was unclear. The mechanism was demonstrated to involve an intermediate species, which was further identified through EXAFS analysis. Here we present a combined XAFS-MCR analysis approach carried out on battery systems to disentangle the electrochemical reaction mechanism. To this purpose, we have selected three case studies: iron hexacyanocobaltate (FeHCC), copper hexacyanoferrate (CuHCF), and vanadium pentoxide (V2O5 ARG-like) cathodes.
3. Results and discussion FeHCC and CuHCF (Mullaliu et al., 2018) belong to the Prussian blue analogues (PBAs) class of materials, which are characterized by a high versatility towards several ions and a low-strain insertion/extraction capability. The EXAFS analysis consisted in this case of a multiple edge refinement, i.e., a simultaneous fitting at both structural
Fig. 1. Concentration profiles (top layer) and pure spectral components (bottom layer) for (a) FeHCC, (b) CuHCF, and (c) V2O5 ARG-like as obtained by the MCR-ALS analysis. The dashed lines in panels a, b separate the charge from the discharge. 2
Radiation Physics and Chemistry xxx (xxxx) xxx–xxx
M. Giorgetti, et al.
by the Department of Energy (DOE) under Contract DE-FG0201ER15221. Measurements at ELETTRA were done through the projects #20130225 and #20145337 (principal investigator MG). Work supported by University of Bologna through RFO funding.
metal edges (Fe, Co for FeHCC and Cu, Fe for CuHCF). The structural information is thus obtained by using two independent measurements, enhancing the reliability and accuracy of the result. On the other side, the MCR-ALS analysis was accomplished only at the Fe and Cu K-edge for FeHCC and CuHCF, respectively, because of the higher variability along the W columns of the experimental XS,W matrix. This can be considered a peculiarity of this class of materials where the metal site linked to the N end of the cyanide group has been found more electroactive than the other metal (linked to the C site) and therefore its spectra during the electrochemical process change dramatically. Despite (i) being structurally very similar displaying both a cubic lattice and a cell parameter of roughly 10 Å, and (ii) being processed in the same way as cathodes for Li-ion batteries, the respective redox mechanisms are different. According to Fig. 1a, FeHCC displays during charge a transformation that can be approximated to the conversion of species 1 to species 2, except for the initial part ascribed to a dehydration process, thus explaining the flat concentration profile (therefore K+ ions are not extracted in this plateau); however, in the following discharge process the system does not react reversibly going back to species 1, instead it evolves to species 3. This fact, though, can be rationalized by the fact that K+ ions are released during charge and Li+ ions are inserted during the subsequent discharge. Unlike, CuHCF material (Fig. 1b), shows a completely reversible interconversion between species 1 and 2 (CuI and CuII, respectively), featuring a 100% relative concentration of species 2 at the end of the charge. This fact can anticipate a good cyclability of this material during the electrochemical performance recording. Fig. 1c highlights the potentiality and the weaknesses of the MCRALS approach while a limited number of spectra is considered. The cathode material was the V2O5 ARG-like which has been studied during discharge, by considering only 25 spectra along the lithiation process. This material belongs to the class of V2O5 gel for lithium intercalation batteries, can be considered a highly interconnected porous material with very thin solid walls and shows a unique lithium intercalation behavior (Coustier et al., 1999). By looking at the concentration profile plot, two exclusive species are gradually interconverted during discharge, being present in equal concentration between spectra 12–13. Eventually, a gradual interconversion of pristine and discharged species has been evidenced, notwithstanding the reduced number of recorded spectra which causes a less clear picture in the concentration profile plot.
References Aquilanti, G., Giorgetti, M., Dominko, R., Stievano, L., Arčon, I., Novello, N., Olivi, L., 2017. Operando characterization of batteries using x-ray absorption spectroscopy: Advances at the beamline XAFS at synchrotron Elettra. J. Phys. D Appl. Phys. 50, 074001. https://doi.org/10.1088/1361-6463/aa519a. Camo Software, The Unscrambler X Patch, v10.2.0.r2. Cassinelli, W.H., Martins, L., Passos, A.R., Pulcinelli, S.H., V Santilli, C., Rochet, A., Briois, V., 2013. Multivariate curve resolution analysis applied to time-resolved synchrotron X-ray Absorption Spectroscopy monitoring of the activation of copper alumina catalyst. Catal. Today 229. https://doi.org/10.1016/j.cattod.2013.10.077. Conti, P., Zamponi, S., Giorgetti, M., Berrettoni, M., Smyrl, W.H., 2010. Multivariate curve resolution analysis for interpretation of dynamic Cu K-edge X-ray absorption spectroscopy spectra for a Cu doped V2O5 lithium battery. Anal. Chem. 82, 3629–3635. https://doi.org/10.1021/ac902865h. Coustier, F., Lee, J.-M., Passerini, S., Smyrl, W.H., 1999. V2O5 aerogel-like lithium intercalation host. Solid State Ionics 116, 279–291. https://doi.org/10.1016/S01672738(98)00349-X. de Juan, A., Tauler, R., 2007. Factor analysis of hyphenated chromatographic data. Exploration, resolution and quantification of multicomponent systems. J. Chromatogr. A 1158, 184–195. https://doi.org/10.1016/j.chroma.2007.05.045. De Juan, A., Jaumot, J., Tauler, R., 2014. Multivariate Curve Resolution (MCR). Solving the mixture analysis problem. Anal. Methods. 6, 4964–4976. https://doi.org/10. 1039/c4ay00571f. Filipponi, A., Di Cicco, A., 1995. X-ray-absorption spectroscopy and n -body distribution functions in condensed matter. II. Data analysis and applications. Phys. Rev. B 52, 15135–15149. https://doi.org/10.1103/PhysRevB.52.15135. Filipponi, A., Di Cicco, A., Natoli, C.R., 1995. X-ray-absorption spectroscopy and n-body distribution functions in condensed matter. I. Theory. Phys. Rev. B 52, 15122–15134. https://doi.org/10.1103/PhysRevB.52.15122. M. Forina, S. Lanteri, C. Armanino, C. Casolino, M. Casale, P. Oliveri, V-PARVUS, (2008). Giorgetti, M., 2013. A review on the structural studies of batteries and host materials by X-ray absorption spectroscopy. ISRN Mater. Sci. 2013, 1–22. https://doi.org/10. 1155/2013/938625. Giorgetti, M., Berrettoni, M., 2008. Structure of Fe/Co/Ni hexacyanoferrate as probed by multiple edge X-ray absorption spectroscopy. Inorg. Chem. 47, 6001–6008. https:// doi.org/10.1021/ic800289c. Giorgetti, M., Berrettoni, M., Filipponi, A., Kulesza, P.J., Marassi, R., 1997. Evidence of four-body contributions in the EXAFS spectrum of Na2Co[Fe(CN)6]. Chem. Phys. Lett. 275, 108–112. https://doi.org/10.1016/S0009-2614(97)00724-0. Goodenough, J.B., Kim, Y., 2010. Challenges for rechargeable Li batteries. Chem. Mater. 22, 587–603. https://doi.org/10.1021/cm901452z. Hong, J., Marceau, E., Khodakov, A.Y., Gaberová, L., Griboval-Constant, A., Girardon, J.S., La Fontaine, C., Briois, V., 2015. Speciation of ruthenium as a reduction promoter of silica-supported co catalysts: a time-resolved in situ XAS investigation. ACS Catal. 5, 1273–1282. https://doi.org/10.1021/cs501799p. Keller, H.R., Massart, D.L., 1991. Peak purity control in liquid chromatography with photodiode-array detection by a fixed size moving window evolving factor analysis. Anal. Chim. Acta 246, 379–390. https://doi.org/10.1016/S0003-2670(00)80976-9. Maeder, M., Zuberbuehler, A.D., 1986. The resolution of overlapping chromatographic peaks by evolving factor analysis. Anal. Chim. Acta 181, 287–291. https://doi.org/ 10.1016/S0003-2670(00)85248-4. Marassi, R., Zamponi, S., Conti, P., Lanteri, S., 2002. Deconvolution of UV-VIS spectroelectrochemical thin-layer data for an EE electrochemical process. Ann. Chim. 92, 261–270. Mullaliu, A., Aquilanti, G., Conti, P., Plaisier, J.R., Fehse, M., Stievano, L., Giorgetti, M., 2018. Copper electroactivity in prussian blue-based cathode disclosed by operando XAS. J. Phys. Chem. C 122, 15868–15877. https://doi.org/10.1021/acs.jpcc. 8b03429. Vandeginste, B.G.M., Massart, D.L., Buydens, L.M.C., De Jong, S., Lewi, P.J., SmeyersVerbeke, J., 1998. Curve and mixture resolution by factor Analysis and related techniques. In: Data Handl. Sci. Technol. Elsevier, pp. 243–306. https://doi.org/10. 1016/S0922-3487(98)80044-0.
4. Conclusions In the reported examples, the XAFS-MCR combination allowed the identification of the different species and the pure spectral components in the explored range of reaction coordinate, which is the potential in an electrochemically driven process such as the one involving a battery charge and discharge. The concentration profile also indicates the reaction paths of the different redox processes. This joint approach has been demonstrated to have a wide potentiality in retrieving the local structural dynamics of selected cathode materials, allowing a quick and efficient data interpretation which would be otherwise time consuming. Acknowledgements Measurements at Brookhaven National Laboratory were supported
3
Contents lists available at ScienceDirect
Radiation Physics and Chemistry journal homepage: www.elsevier.com/locate/radphyschem
XAFS studies on battery materials: Data analysis supported by a chemometric approach Marco Giorgettia,∗, Angelo Mullaliua, Paolo Contib a b
Department of Industrial Chemistry “Toso Montanari”, University of Bologna, Viale Risorgimento 4, 40136, Bologna, Italy School of Science, Chemistry Division, University of Camerino, 62032, Camerino, Italy
A R T I C LE I N FO
A B S T R A C T
Keywords: XAFS EXAFS MCR-ALS Li-ion battery Operando
The large amount of data collected during a typical operando X-ray absorption fine structure (XAFS) experiment and the interest to thoroughly investigate and comprehend a redox process occurring in a Li-ion battery need an advanced data analysis approach. In this context, multivariate curve resolution (MCR) has gained attention as chemometric technique that allows the identification of the number of the species and their existence range without any a priori information on the studied system. The call for an advanced data treatment is answered by a joint XAFS-MCR approach, able to unveil the local structural dynamics of the selected species, providing a quick and efficient data interpretation.
1. Introduction The development of reliable, low-cost, sustainable storage of electrical energy generated by sources other than fossil fuels is a global priority. Rechargeable batteries, which are based on high rate intercalation reaction of lithium ions into nano- and micro-structured porous materials, offer efficient electrochemical energy storage (Goodenough and Kim, 2010). The structural and electronic evolution of such hosts is of fundamental importance to design advanced tailored materials, making X-ray absorption fine structure (XAFS) the technique of choice for retrieving this information (Giorgetti, 2013). During a typical operando XAFS experiment, a large amount of data can be acquired, calling for a suitable strategy to data treatment: to this purpose, a joint XAFS-chemometric approach on large datasets allows an efficient and in-depth investigation of the (de)lithiation process. By recording the electrochemical response relatively slowly compared to the spectral data collection, it is possible to build an experimental data matrix (XS,W) where each row is a XAFS spectrum in a very narrow potential range. Multivariate curve resolution refined by alternating least squares (MCR-ALS) provides an additive bilinear model of pure contributions without any preexisting model or a priori information on the system, decomposing the experimental matrix in a product of two matrices to which chemical meaning can be attributed (De Juan et al., 2014; Conti et al., 2010), as it follows (Eqn. (1)):
X s,w = Cs,f Af,w + Es,w ∗
(1)
where CS,F is the column matrix of the concentration profile, AF,W is the row matrix of the X-ray absorption near edge structure (XANES) spectra of pure species and ES,W is the matrix of the residue. A tentative initial value, as close as possible to the final solution, of one of the matrices must be provided for initializing the ALS refinement. A decomposition of the experimental matrix similar to that in Eqn. (1) can be done by Principal component analysis (PCA), however, the two computed matrices do not have physico-chemical meaning, being the main result of PCA the rank of the matrix XS,W that, in the present example, would correspond to the number of pure species. Evolving factor analysis (EFA) (Maeder and Zuberbuehler, 1986) and fixed size window EFA (FSWEFA) (Keller and Massart, 1991) methods (Vandeginste et al., 1998) are preferred in place of PCA because they are able of guessing both the number of pure species and a first estimate of their concentration profiles that are useful to initialize the refinement (Vandeginste et al., 1998; Marassi et al., 2002). Moreover, to obtain a unique and meaningful solution, constraints such as the closure condition and non-negativity (of concentration and/or of pure response) must be applied. A logical tree diagram (Scheme 1) is reported to visualize the steps involved in the MCR-ALS analysis. MCR-ALS technique has been applied in several fields, for instance in chromatography (de Juan and Tauler, 2007), and on in situ XAFS data on catalysis (Cassinelli et al., 2013; Hong et al., 2015). The first example of a joint XAFS-MCR approach dates back to 2010 (Conti et al., 2010), when Conti et al. studied Cu-doped vanadium pentoxide (Cu0.1V2O5) for its attractive properties as insertion host for Li-ions
Corresponding author. E-mail address: [email protected] (M. Giorgetti).
https://doi.org/10.1016/j.radphyschem.2019.04.002 Received 21 September 2018; Received in revised form 14 November 2018; Accepted 1 April 2019 Available online 04 April 2019 0969-806X/ © 2019 Elsevier Ltd. All rights reserved.
Please cite this article as: Marco Giorgetti, Angelo Mullaliu and Paolo Conti, Radiation Physics and Chemistry, https://doi.org/10.1016/j.radphyschem.2019.04.002
Radiation Physics and Chemistry xxx (xxxx) xxx–xxx
M. Giorgetti, et al.
2. Experimental X-ray absorption spectroscopy experiments for FeHCC and CuHCF cathodes were performed at the XAFS beamline of Elettra Sincrotrone in Basovizza (Italy). (Aquilanti et al., 2017). Data collection was carried out during the first charge/discharge cycle in the 1.8 < E < 4.0 V and 2.0 < E < 4.3 V vs. Li+/Li, respectively. The data were recorded in transmission mode either at Fe, Co and/or Cu K-edges. The energies were defined by assigning the first inflection point of the spectra of the metallic foil of Fe, Co and/or Cu to 7112, 7709 and 8979 eV, respectively. An internal reference of Fe, Co and/or Cu foil was used for energy calibration in each scan. Experiments for V2O5 ARG-like were performed at the V K-edge at the National Synchrotron Light Source (NSLS) at Brookhaven National Laboratory using the beamline X11 A (Conti et al., 2010). Data collection was carried out in transmission mode while discharging from open voltage circuit (OCP) value (3.3 V) to 1.5 V vs. Li+/Li. The energy was defined by assigning the first inflection point of the spectra of a vanadium foil. The extended X-ray absorption fine structure (EXAFS) analysis was performed by using the GNXAS package (Filipponi et al., 1995; Filipponi and Di Cicco, 1995) that takes into account the multiple scattering (MS) theory. The theoretical signal is calculated ab initio and contains the relevant two-body γ(2), three-body γ(3), and four-body γ(4) MS terms, (Giorgetti et al., 1997). A typical structure of hexacyanoferrates (Giorgetti and Berrettoni, 2008) was used as structural model for the FeHCC and CuHCF MS signals calculation. The MCR-ALS analysis was performed by taking into account the XANES portion of the XAFS spectrum exclusively. The Unscrambler (Camo Software) and homemade routines based on Parvus (Forina et al., 2008) was used in data treatment.
Scheme 1. MCR-ALS analysis diagram.
batteries. Although the compound was known since years among the scientific community, the actual reaction mechanism occurring during charge was unclear. The mechanism was demonstrated to involve an intermediate species, which was further identified through EXAFS analysis. Here we present a combined XAFS-MCR analysis approach carried out on battery systems to disentangle the electrochemical reaction mechanism. To this purpose, we have selected three case studies: iron hexacyanocobaltate (FeHCC), copper hexacyanoferrate (CuHCF), and vanadium pentoxide (V2O5 ARG-like) cathodes.
3. Results and discussion FeHCC and CuHCF (Mullaliu et al., 2018) belong to the Prussian blue analogues (PBAs) class of materials, which are characterized by a high versatility towards several ions and a low-strain insertion/extraction capability. The EXAFS analysis consisted in this case of a multiple edge refinement, i.e., a simultaneous fitting at both structural
Fig. 1. Concentration profiles (top layer) and pure spectral components (bottom layer) for (a) FeHCC, (b) CuHCF, and (c) V2O5 ARG-like as obtained by the MCR-ALS analysis. The dashed lines in panels a, b separate the charge from the discharge. 2
Radiation Physics and Chemistry xxx (xxxx) xxx–xxx
M. Giorgetti, et al.
by the Department of Energy (DOE) under Contract DE-FG0201ER15221. Measurements at ELETTRA were done through the projects #20130225 and #20145337 (principal investigator MG). Work supported by University of Bologna through RFO funding.
metal edges (Fe, Co for FeHCC and Cu, Fe for CuHCF). The structural information is thus obtained by using two independent measurements, enhancing the reliability and accuracy of the result. On the other side, the MCR-ALS analysis was accomplished only at the Fe and Cu K-edge for FeHCC and CuHCF, respectively, because of the higher variability along the W columns of the experimental XS,W matrix. This can be considered a peculiarity of this class of materials where the metal site linked to the N end of the cyanide group has been found more electroactive than the other metal (linked to the C site) and therefore its spectra during the electrochemical process change dramatically. Despite (i) being structurally very similar displaying both a cubic lattice and a cell parameter of roughly 10 Å, and (ii) being processed in the same way as cathodes for Li-ion batteries, the respective redox mechanisms are different. According to Fig. 1a, FeHCC displays during charge a transformation that can be approximated to the conversion of species 1 to species 2, except for the initial part ascribed to a dehydration process, thus explaining the flat concentration profile (therefore K+ ions are not extracted in this plateau); however, in the following discharge process the system does not react reversibly going back to species 1, instead it evolves to species 3. This fact, though, can be rationalized by the fact that K+ ions are released during charge and Li+ ions are inserted during the subsequent discharge. Unlike, CuHCF material (Fig. 1b), shows a completely reversible interconversion between species 1 and 2 (CuI and CuII, respectively), featuring a 100% relative concentration of species 2 at the end of the charge. This fact can anticipate a good cyclability of this material during the electrochemical performance recording. Fig. 1c highlights the potentiality and the weaknesses of the MCRALS approach while a limited number of spectra is considered. The cathode material was the V2O5 ARG-like which has been studied during discharge, by considering only 25 spectra along the lithiation process. This material belongs to the class of V2O5 gel for lithium intercalation batteries, can be considered a highly interconnected porous material with very thin solid walls and shows a unique lithium intercalation behavior (Coustier et al., 1999). By looking at the concentration profile plot, two exclusive species are gradually interconverted during discharge, being present in equal concentration between spectra 12–13. Eventually, a gradual interconversion of pristine and discharged species has been evidenced, notwithstanding the reduced number of recorded spectra which causes a less clear picture in the concentration profile plot.
References Aquilanti, G., Giorgetti, M., Dominko, R., Stievano, L., Arčon, I., Novello, N., Olivi, L., 2017. Operando characterization of batteries using x-ray absorption spectroscopy: Advances at the beamline XAFS at synchrotron Elettra. J. Phys. D Appl. Phys. 50, 074001. https://doi.org/10.1088/1361-6463/aa519a. Camo Software, The Unscrambler X Patch, v10.2.0.r2. Cassinelli, W.H., Martins, L., Passos, A.R., Pulcinelli, S.H., V Santilli, C., Rochet, A., Briois, V., 2013. Multivariate curve resolution analysis applied to time-resolved synchrotron X-ray Absorption Spectroscopy monitoring of the activation of copper alumina catalyst. Catal. Today 229. https://doi.org/10.1016/j.cattod.2013.10.077. Conti, P., Zamponi, S., Giorgetti, M., Berrettoni, M., Smyrl, W.H., 2010. Multivariate curve resolution analysis for interpretation of dynamic Cu K-edge X-ray absorption spectroscopy spectra for a Cu doped V2O5 lithium battery. Anal. Chem. 82, 3629–3635. https://doi.org/10.1021/ac902865h. Coustier, F., Lee, J.-M., Passerini, S., Smyrl, W.H., 1999. V2O5 aerogel-like lithium intercalation host. Solid State Ionics 116, 279–291. https://doi.org/10.1016/S01672738(98)00349-X. de Juan, A., Tauler, R., 2007. Factor analysis of hyphenated chromatographic data. Exploration, resolution and quantification of multicomponent systems. J. Chromatogr. A 1158, 184–195. https://doi.org/10.1016/j.chroma.2007.05.045. De Juan, A., Jaumot, J., Tauler, R., 2014. Multivariate Curve Resolution (MCR). Solving the mixture analysis problem. Anal. Methods. 6, 4964–4976. https://doi.org/10. 1039/c4ay00571f. Filipponi, A., Di Cicco, A., 1995. X-ray-absorption spectroscopy and n -body distribution functions in condensed matter. II. Data analysis and applications. Phys. Rev. B 52, 15135–15149. https://doi.org/10.1103/PhysRevB.52.15135. Filipponi, A., Di Cicco, A., Natoli, C.R., 1995. X-ray-absorption spectroscopy and n-body distribution functions in condensed matter. I. Theory. Phys. Rev. B 52, 15122–15134. https://doi.org/10.1103/PhysRevB.52.15122. M. Forina, S. Lanteri, C. Armanino, C. Casolino, M. Casale, P. Oliveri, V-PARVUS, (2008). Giorgetti, M., 2013. A review on the structural studies of batteries and host materials by X-ray absorption spectroscopy. ISRN Mater. Sci. 2013, 1–22. https://doi.org/10. 1155/2013/938625. Giorgetti, M., Berrettoni, M., 2008. Structure of Fe/Co/Ni hexacyanoferrate as probed by multiple edge X-ray absorption spectroscopy. Inorg. Chem. 47, 6001–6008. https:// doi.org/10.1021/ic800289c. Giorgetti, M., Berrettoni, M., Filipponi, A., Kulesza, P.J., Marassi, R., 1997. Evidence of four-body contributions in the EXAFS spectrum of Na2Co[Fe(CN)6]. Chem. Phys. Lett. 275, 108–112. https://doi.org/10.1016/S0009-2614(97)00724-0. Goodenough, J.B., Kim, Y., 2010. Challenges for rechargeable Li batteries. Chem. Mater. 22, 587–603. https://doi.org/10.1021/cm901452z. Hong, J., Marceau, E., Khodakov, A.Y., Gaberová, L., Griboval-Constant, A., Girardon, J.S., La Fontaine, C., Briois, V., 2015. Speciation of ruthenium as a reduction promoter of silica-supported co catalysts: a time-resolved in situ XAS investigation. ACS Catal. 5, 1273–1282. https://doi.org/10.1021/cs501799p. Keller, H.R., Massart, D.L., 1991. Peak purity control in liquid chromatography with photodiode-array detection by a fixed size moving window evolving factor analysis. Anal. Chim. Acta 246, 379–390. https://doi.org/10.1016/S0003-2670(00)80976-9. Maeder, M., Zuberbuehler, A.D., 1986. The resolution of overlapping chromatographic peaks by evolving factor analysis. Anal. Chim. Acta 181, 287–291. https://doi.org/ 10.1016/S0003-2670(00)85248-4. Marassi, R., Zamponi, S., Conti, P., Lanteri, S., 2002. Deconvolution of UV-VIS spectroelectrochemical thin-layer data for an EE electrochemical process. Ann. Chim. 92, 261–270. Mullaliu, A., Aquilanti, G., Conti, P., Plaisier, J.R., Fehse, M., Stievano, L., Giorgetti, M., 2018. Copper electroactivity in prussian blue-based cathode disclosed by operando XAS. J. Phys. Chem. C 122, 15868–15877. https://doi.org/10.1021/acs.jpcc. 8b03429. Vandeginste, B.G.M., Massart, D.L., Buydens, L.M.C., De Jong, S., Lewi, P.J., SmeyersVerbeke, J., 1998. Curve and mixture resolution by factor Analysis and related techniques. In: Data Handl. Sci. Technol. Elsevier, pp. 243–306. https://doi.org/10. 1016/S0922-3487(98)80044-0.
4. Conclusions In the reported examples, the XAFS-MCR combination allowed the identification of the different species and the pure spectral components in the explored range of reaction coordinate, which is the potential in an electrochemically driven process such as the one involving a battery charge and discharge. The concentration profile also indicates the reaction paths of the different redox processes. This joint approach has been demonstrated to have a wide potentiality in retrieving the local structural dynamics of selected cathode materials, allowing a quick and efficient data interpretation which would be otherwise time consuming. Acknowledgements Measurements at Brookhaven National Laboratory were supported
3